Tencent/TurboTransformers

Refactor MultiheadAttention and other Layers

オープン

#140 opened on 2020/07/22

 (0 件のコメント) (0 件のリアクション) (0 人の担当者)C++ (206 件のフォーク)github user discovery
enhancementhelp wanted

Repository metrics

Stars
 (1,546 個のスター)
PR merge metrics
 (PR metrics pending)

説明

Now the logic inside MultiheadAttention Layer is too complex for development. Moreover, some bugs exist in intermediate management. It is the first priority to rewrite these codes to make others easily understand what Turbo is doing.

コントリビューターガイド